# Julien Wuthrich: selfjev

> An independent open decision model with Jev's request format: a rank-64 LoRA on Qwen3.5-4B that reads text or an image once and scores yes/no, one-choice, ordered-score and multi-select questions from forward-pass logits, returning per-option probabilities.

- Page: https://systemonemodels.ai/julien-wuthrich/selfjev
- API: https://api.systemonemodels.ai/v1/models/julien-wuthrich/selfjev
- Download: `pip install systemonemodels && systemone pull julien-wuthrich/selfjev`

## Facts

| | |
|---|---|
| Maker | Julien Wuthrich (https://systemonemodels.ai/julien-wuthrich) |
| Decides | noul, choice, score, classify, route |
| Architecture | selfjev |
| Base model | qwen/qwen3.5-4b |
| Parameters | 4.0B |
| Licence | Weights: no licence declared, treated as non-commercial; code Apache-2.0 |
| Availability | Open weights |
| Released | 2026-09-30 |
| Latest version | 2026.09 |

## Reported evaluation

Suite: selfjev-decision-bench Text Decisions (1,991 questions, AI-checked authored labels; the maker's suite). Numbers are the publisher's own.

- Decision accuracy: 96.1%

## Model card

<!-- generated by scripts/seed_catalog.py; edit content/models/catalog.yaml -->

# SelfJev-4B Vision

An independent open decision model with Jev's request format: a rank-64 LoRA on Qwen3.5-4B that reads text or an image once and scores yes/no, one-choice, ordered-score and multi-select questions from forward-pass logits, returning per-option probabilities.

A LoRA adapter of about 230 MB; the Qwen3.5-4B base downloads separately. Served by the maker's selfjev package, which shares the prefix across questions and candidates; merged checkpoints also run on vLLM, but images need the maker's server. It accepts Jev-shaped requests, and the maker says code written for TypeSafe's SDK runs unchanged. Half the training targets are Jev API probabilities and half AI-checked labels. Not calibrated, per the maker. No licence is declared for the fine-tuned weights; the code is Apache-2.0. The maker's comparisons are on its own suites, which it says favour SelfJev. Siblings: the text-only selfjev-4b, merged checkpoints and GGUF builds.

## What it decides

- **noul** — answers a yes/no question with one probability
- **choice** — picks one option from a set
- **score** — places the input on an ordered scale
- **classify** — assigns a category from a fixed taxonomy
- **route** — sends the input to one of several destinations

## At a glance

| | |
|---|---|
| Parameters | 4B |
| Base model | `Qwen/Qwen3.5-4B` |
| Maker | Julien Wuthrich |
| Released | 2026-09-30 |
| License | Weights: no licence declared, treated as non-commercial; code Apache-2.0 |
| Reported accuracy | 96.1% |
| Reported latency | 73 ms median, one text question, one warm request on an NVIDIA L40S |

## Get the weights

```bash
pip install systemonemodels
systemone pull julien-wuthrich/selfjev
```

The files are served from the maker's Hugging Face repository, [`Jwuthrich/selfjev-4b-vision`](https://huggingface.co/Jwuthrich/selfjev-4b-vision), and verified against the checksums recorded here.

## Read more

- [Model card](https://huggingface.co/Jwuthrich/selfjev-4b-vision)
- [Text-only adapter](https://huggingface.co/Jwuthrich/selfjev-4b)
- [Source code](https://github.com/Jwuthri/SelfJev)

---

*This page was opened by System One for Julien Wuthrich, who can claim the organisation and take it over at any time.*

---

From System One Models — https://systemonemodels.ai/ · every System One model: https://systemonemodels.ai/system-one-models